Data Infrastructure Engineer – Exabyte-Scale Platform
Data Infrastructure Engineer at Mistral AI, building and operating exabyte-scale data lakes, storage, and multi-cluster compute orchestration (Kubernetes/SLURM) that power frontier model training and fine-tuning. Work spans Python, columnar storage formats, metadata/lineage systems, and on-call for critical training jobs.
- Compensation
- Not disclosed
- Location
- Paris, France
- Work setup
- Not specified
- Seniority
- Not specified
- Employment
- Not specified
- Experience
- 4+ years
- Category
- Devops
- Posted
Skills
Must have
- 4+ years of experience in Data Infrastructure, MLOps, or Infrastructure Engineering
- Experience or strong interest in supporting foundational compute and storage platforms
- Proficient in Python; experience with modern columnar storage standards
- Well-versed in Kubernetes-native tooling and multi-cluster environments
- Experience debugging large-scale distributed systems
- Able to build and operate scalable, reliable, and secure systems from the ground up
- Comfortable with ambiguity in a rapid-growth AI environment
Nice to have
- Experience with model training and fine-tuning platforms across Kubernetes and SLURM environments
- Based in or open to relocating to Paris, London, Warsaw, or Zurich
Full description
About Mistral
Mistral provides full-stack AI solutions: from frontier models to developer tools, applications, and compute. We partner with enterprises tackling the hardest problems across high-stakes industries like finance, manufacturing, defense, healthcare, and the public sector, co-creating customized AI systems that they can run on their terms.
We are a dynamic, collaborative team passionate about AI and its potential to transform society. Our diverse workforce thrives in competitive environments and is committed to driving innovation. Our teams are distributed between Europe, North America, Asia and the Middle East. We are creative, low-ego and team-spirited.
Role Summary
The Data Infrastructure team at Mistral AI is architecting the backbone of our frontier model training and fine-tuning ecosystem. We are building the specialized compute and data fabrics required to power the development of world-class AI.
Our vision is to operate some of the largest compute fleets in production and build data lakes and metadata systems with a roadmap toward exabyte-scale architecture. We are currently in the process of building a high-performance training platform designed for massive scale across both on-premise and cloud-native Kubernetes environments.
We are leading a strategic transition from legacy scheduling to modern orchestration. With numerous clusters distributed across various regions, we are focussed on implementing sophisticated multi-cluster orchestration and cloud-bursting capabilities to better utilize our global resources and ensure our researchers have seamless access to compute wherever it resides. Our mission is to evolve our current systems into a platform that is as durable as it is flexible.
Location: Paris / Warsaw / Zurich / London (hybrid) or remote EU/UK with one hub visit per month.
About the Role
This role focuses on building and operating the next generation of data infrastructure at Mistral AI. You will be a core contributor to our evolution, helping us design and scale massive compute fleets and storage systems designed for high performance and scalability.
You will help us move toward a future of decoupled control and data planes, scaling big data compute and storage platforms while ensuring secure and governed data access for MLOps and research. You will take full lifecycle ownership: from architecting the migration away from legacy orchestrators to implementing production-grade pipelines and participating in on-call rotations for critical training jobs.
In this role, you will:
- Build & Scale: Help us reach our goal of operating massive distributed compute and storage systems
- Global Orchestration: Architect and maintain multi-cluster orchestration layers to optimize workload placement across diverse hardware and regions.